Fischer-Tropsch, organometallics, and other friends

Recent researches on the mechanism of the Fischer-Tropsch hydrogenation of CO heterogeneously catalyzed by ruthenium, rhodium, cobalt or iron (on silica), both with and without added probe molecules, are reviewed. When the reactions are carried out under mild conditions of temperature, pressure and catalyst, we find that, although the absolute activities of the four catalysts differ, the relative product distributions are rather similar. Our data indicate that the primary products are largely n-l-alkenes (and methane); n-alkanes and some internal olefins are also produced, but these arise substantially from subsequent (secondary) reactions of the 1-alkenes. Mechanisms for the formation of 1-alkenes are suggested, both from our work and that of other groups, based upon data from (i) a variety of labeling and spectroscopic studies, (ii) model stoichiometric reactions of organometallic complexes, and (iii) some theoretical and computational data now available. The mechanism in best agreement with experimental data involves, (a) initiation via deoxygenation of coordinated CO and the production of a C-2 based surface species, probably a vinyl {CH2=CH(ad)}, via formation of surface carbide, methylidyne and methylene; (b) propagation by reaction of surface methylene (CH2(ad)) with surface vinyl or alkenyl {RCH=CH(ad)} to give a surface allyl {RCH=CHCH2(ad)}, followed by a 1,3-H shift to generate a new surface alkenyl {RCH2CH=CH(ad)}; and (c) termination by hydride-mediated cleavage of the alkenyl chain from the catalyst giving the 1-alkene, RCH2CH=CH2, directly. By contrast, the hydrogenation of CO homogeneously catalyzed by soluble complexes (chiefly of Ru) is reported to occur only under morestringent conditions and to lead mainly to methanol and ethylene glycol; clearly quite different mechanisms operate there. (C) 2004 Elsevier B.V. All rights reserved.